Support is requested for research on the structure, stability, and mechanism of interaction with DNA and proteins of HMG box proteins, members of a family of DNA binding proteins active in DNA packing, control of transcription, and recombinational repair mechanisms. The planned experiments will investigate the binding properties of single copy boxes, two copies in tandem, and a version of the latter including the strongly acidic C terminal tail sequence. Two boxes associate less strongly with DNA than single boxes; the mechanism of this anti- cooperative effect will be investigated in detail. Complexes of wild type and mutant proteins with a variety of branched DNA substrates, including short chain three and four way DNA junctions, and longer branched substrates formed from synthetic polynucleotide copolymers and circular DNAs, will be investigated. Methods to be used include reaction and scanning calorimetry, CD and fluorescence spectroscopy, topographical protein mapping, and differential hydrogen exchange rate measurements. One particular objective of the project is to attempt to assemble a quantitative picture of the contribution of specific side chain interactions in stabilizing the complex between HMG's and branched or bent DNA, an "interaction map" of the complex. Mutant and wild type protein sequences will be compared to identify the distinct requirements for folding and protein-protein interaction from those for DNA binding. The process whereby high molecular DNA is progressively wrapped or folded into loops in the presence of multiple HMG proteins will also be studied, using atomic force microscopy and protein topographical mapping. Fragments of the proteins will be investigated in order to identify functional subunits in HMG boxes. These will be synthesized as peptides and studied in competition experiments to define their capabilities.